STAGE 5 - Data Analysis and Delivery
The key to performance indicators is interpretation. By understanding what the data is for, and the environment in which it is collected, we can apply several modifications to convert the data into useful knowledge.
There are occasions where underlying regional trends need to be rectified. For example funding information‐different global currencies must of course be converted to a common currency. However there are also different costs for living, running a university, and conducting research. Such differences in costs can make direct funding comparisons difficult, and we used Purchasing Power Parity to overcome the regional variance and make data comparable.
Another important modification to data is normalization for institution size. Without this modification the data is purely a measure of volume and has little value as performance indicators. For example, if we are to look at the number of scholarly articles published by a university, there is a clear indication of volume and we may find that some of the larger institutions publish the most simply because of their size. However, if we look at the number of articles published per person, this indicator becomes a more meaningful measure of the university’s effectiveness rather than its overall output.
Even after these modifications have been made, the value of a single indicator is a largely meaningless number unless it is clear where that value falls within its surrounding landscape. It is by understanding how a value compares to its peers and where it fits into a distribution of other institutions that we can convert the raw data into actionable information. For example, consider an institution that has a ratio of doctoral students to undergraduate students of 0.3. Without a reference point, it is difficult to tell if 0.3 is a high or low value. One way in which you might interpret this number is to benchmark the value against peer groups,i.e., compare the value to that of the average for a similar group of universities. This is exactly the type of analysis that we use to offer clearer pictures of how universities actually perform. In the example given, the average of all universities in the Profiles Project is 0.15, so one can quickly draw the conclusion that the institution performs at twice the level when compared to similar institutions.
As a user of the profiles data you might be interested to see how an indicator compares to a different peer group, for example a group of prestigious universities or a group of universities with a similar subject focus. Alternatively, you might want to see how an indicator varies within the different subject areas at an institution and how each subject area compares to a group of peers. All of these types of analysis are feasible and can lead to some very informative profiles and reports.
PROFILES PROJECT – MOVING FORWARD
The first use of the data generated in the Global Institutional Profiles Project was to inform the Times Higher Education World University Ranking. However, there are many other services that will rely on the Profiles Project data. For example the data can be used to inform customized analytical reporting or customized data sets for a specific customer’s needs.
Thomson Reuters is developing a platform designed for easy access and interpretation of this valuable data set. The platform will combine different sets of key indicators, with peer benchmarking and visualization tools to allow users to quickly identify the key strengths of institutions across a wide variety of aspects and subjects.